661. Vagus nerve stimulation reduces suicidal ideation in markedly treatment-resistant depression: RECOVER findings

Abstract

Abstract Background Individuals with markedly treatment-resistant depression (mTRD) have substantial risk for suicide attempt and completion. Few antidepressant treatments offer benefit for suicide, and it is critical that novel treatments not increase/worsen suicidal ideation (SI) or behavior. Aims & Objectives Assess the impact of adjunctive vagus nerve stimulation (VNS) on SI in patients with mTRD. Method Participants (N=493) enrolled in RECOVER, a 12-month, triple-blinded, controlled trial were randomized 1:1 to receive active or sham VNS adjunctive with treatment as usual. Suicide assessments were performed monthly from months 3 to 12 (10-month observation period). Patients with active SI or a suicide attempt in the previous 6 months were excluded. The sample was profoundly treatment resistant, chronically ill, and disabled, with an average of >13 failed lifetime antidepressant treatments and a current major depressive episode duration of nearly 18 years; 75% were unemployed. Approximately 40% with major depressive disorder had a previous history of suicide attempt, and ~70% entered the trial with ongoing SI. The outcome measure was a Composite Suicidal Ideation (CSI) score combining the suicide items on the blinded, off-site, clinician-rated Montgomery-Åsberg Depression Rating Scale (MADRS) and Quick Inventory of Depressive Symptomatology–Clinician (QIDS-C), and the patient-rated Quick Inventory of Depressive Symptomatology–Self-Report (QIDS-SR). The total CSI score ranged from 0 to 9. Analyses were conducted separately on those with clinically meaningful baseline SI (defined as a CSI score ≥3) and on those with minimal or no baseline SI. Assessments determined whether the treatment groups differed in the proportion of participants who showed meaningful reduction in SI (reduction of ≥3 from baseline CSI score) over the 10 assessment occasions (months 3–12) and during the final 3 months (months 10–12), a time-frame during which maximal benefit of VNS was previously observed. Remission of SI (defined as a reduction of ≥3 points and a CSI score of ≤2) and emergence of clinically meaningful SI were also assessed. Results Of the 493 patients entering the trial, 463 had ≥1 suicidal assessment after implantation. Among those with significant SI at baseline (N=286), as compared to sham VNS, those receiving active VNS had significantly greater clinically meaningful reduction in SI over the 10-month observation period (odds ratio [OR], 1.43; P=0.047) with greater reductions observed from the first assessment onward. Across months 10–12, active VNS showed statistically superior reduction in likelihood of remission of SI compared to sham VNS (OR, 1.67; P=0.047). Across all subjects, emergence of SI was rare, and active VNS (as compared to sham) was not associated with emergence of greater SI in either those with (active VNS, 1.4%; sham VNS, 1.6%) or without (active VNS, 4.4%; sham VNS, 3.6%) clinically meaningful SI at baseline. Discussion & Conclusions During the blinded and randomized phase of the study, active VNS resulted in greater reductions in SI than sham VNS. Further, as compared to sham VNS, active VNS did not result in worsening SI. VNS is safe in patients with markedly TRD with SI and could potentially contribute to reduction in SI.